PROJECT SUMMARY/ABSTRACT Non-adherence to antiepileptic drugs (AEDs) is a common problem (i.e., 58% of patients have some level of non-adherence) for young children with newly diagnosed epilepsy, with potentially devastating consequences. AED non-adherence is associated with a 3-fold increased risk of seizures, poor quality of life, inaccurate clinical decision-making, and higher health care utilization and costs. One of the primary barriers to adherence is forgetting, which may be particularly amenable to mHealth (mobile technology in healthcare) interventions. Despite the critical need to develop and implement interventions to improve adherence, there are few family- based interventions for young children with epilepsy and their families, with the exception of the PI's pilot and existing R01 trial. Although highly promising, this intervention requires six in-person sessions, which can be impossible for families who lack routine access to tertiary specialty care due to time, financial, or transportation constraints. Thus, unmet medical and psychosocial needs of the underserved pediatric epilepsy population are perpetuated and compounded by limited access to this state of the art care. Our overall goal is to test a mHealth adherence intervention that is easily accessible using a stepped up care model based on individual needs. This stepped up care model will conserve patient, family, and provider time, costs and resources. The aim of this multi-site R01 is to conduct a two-stage, sequential, multiple assignment, randomized trial (SMART) to evaluate the effectiveness of mHealth intervention strategies for improving AED adherence in caregivers of young children with epilepsy. A two-month baseline period will be followed by two stages. In Stage 1 (3-months long), non-adherent caregivers (< 95%) will be randomized to a mHealth education module and automated digital reminders (control) or the mHealth education module, automated digital reminders, and individualized adherence feedback based on real-time adherence monitoring (treatment) to address the primary barrier of forgetting. At the beginning of Stage 2 (two months long), caregivers randomized to treatment who do not achieve adherence > 95% (response) by the end of Stage 1 will be re-randomized to either continued individualized adherence feedback or individualized adherence feedback augmented with two mHealth problem-solving modules (translated from the PIs existing RCTs) with a therapist. Thus, there are three intervention strategies embedded in this SMART: #1 control, #2 treatment, and #3 problem-solving augmented treatment if nonresponsive at three months. The primary outcome is electronically-monitored adherence and secondary outcomes include seizure severity/frequency, quality of life, and healthcare utilization. If the aims of the project are achieved, this study would have a large impact on pediatric epilepsy, with the potential to change clinical practice for treating non-adherence. The SMART design would allow us to identify patients who are most likely to respond to interventions and step up care with more time- and resource-intensive interventions (i.e., problem-solving with a therapist via the web), when necessary.